In recent months, the amount of innovation coming out of the OpenSim community has been staggering. Hypergrid travel is not only more secure, but now supports cross-grid friendships, messages, and landmarks. OpenSim providers are hosting their regions on-demand and in public (and private) clouds, offering full regions for under $10 a month, allowing users to sign in with Facebook accounts and even access the grids from within Facebook pages. There are now two multi-grid content marketplaces — HGExchange and Cariama — and a number of Websites offering OpenSim content.
OpenSim has had mesh for over three years with realXtend and the modrex module — and Second Life-compatible mesh is now standard on every grid running the latest version of OpenSim. In fact, when Second Life added mesh support to its experimental viewer last fall, OpenSim was supporting it the next day.
And OpenSim grids and developers are experimenting with new physics and scripting engines, allowing people to host regions at home for free while still being part of a bigger grid, and — oh yeah — have figured out how to have over 1,000 simultaneous visitors on a single region.
The total number of users of any OpenSim grid is tiny compared to Second Life, but growth is accelerating. And if the pace of innovation continues, Second Life is going to be facing some serious competition very, very soon.
So why can’t Linden Lab innovate? Why did it take more than two years for Second Life to get mesh after OpenSim did? The biggest recent improvement is with display names — while OpenSim grids have been allowing users to pick any names they wanted right from the start.
Why are prices so high and service so bad that popular and heavily marketed in-world virtual pets almost starved to death this week?
Or read on right here:
1. Legacy technology
Linden Lab has a great deal of time and money invested in its proprietary technology. Technology that was created years ago, and is clunky and unwieldy compared to the more up-to-date, modular OpenSim software.
That means that OpenSim developers can easily plug in new modules that dramatically extend the way OpenSim works. They can plug in new physics engines. Add mesh support. Add new scripting commands. Add different currency and payment modules.
Pretty much anything they want — all without changing the core software.
Over time, OpenSim developers are continuing to streamline the system so that modules are integrated even more smoothly, and can access even more of the functionality of OpenSim.
Meanwhile, with the Second Life software, each new change has to be integrated with all the code that was written before — meaning that each successive innovation takes even longer to write, implemented and debug than the one before.
2. Legacy costs
An OpenSim hosting company can set up shop and offer regions for $60 a month. For $9.90 a month. Even for 10 cents a month.
After all, the software has already been written, and is available for free. And typical startup’s overhead costs are low — in fact, the founders often work for free. And by using rented or cloud-based infrastructure, a grid doesn’t have to pay for any hardware costs up front, and just add on as needed.
Linden Lab, by comparison, has a huge payroll to meet each money. High-priced senior executives to keep happy. And data centers full of servers to keep powered up.
It can’t just decide to release a low-cost version of its product. If it did, and all customers switched over — it would be out of business overnight.
This is similar to the problem that newspapers are having with the Internet. On the one hand, they’re charging people for paper subscriptions. On the other hand, all the same news — and, often, even more — is available online for free. But there’s not enough online revenues to support the news organization yet. The result? Print publications are folding right and left while startup online news sites are opening up everywhere.
In The Innovator’s Dilemma, Christensen tracks several such cycles in the hard disk industry. New companies kept coming out with cheaper, better technologies — and the old companies would go out of business because they couldn’t transition with their existing business structures.
The rise of the Internet meant that every single industry was now faced with a similar dilemma, and the process continues to this day. Blockbuster couldn’t compete with Netflix. Borders couldn’t compete with Amazon. Banks couldn’t figure out how to send money over the Internet, letting PayPal dominate the online payments market.
AOL lost its leading position to upstarts like Yahoo and Google — and now makes its money from people who don’t need its services anymore but don’t know how to cancel their accounts.
3. Legacy customers
I’m not saying that there’s anything wrong with Second Life’s existing customers. Just that the reason they are customers is that they like things they way they are. Otherwise, they wouldn’t be customers. And they’re willing to pay $300 per region per month. These are valuable, valuable customers.
The problem is that customers like thing they way they are — and then, suddenly, they don’t. There’s no warning.
Consider customer relationship management software. Companies spends millions on CRM, on integrating it with their systems, teaching their employees to use it. It holds their most valuable information. The last thing they want to do is let some cheap, insecure startup touch it. So, at the beginning, none of the big CRM vendors were worrying about Salesforce.com’s cheap, web-based CRM. After all, none of their customers wanted it. The people who wanted it were tiny, dinky little companies that didn’t have any money to spend anyway — who cared about them?
But it turned out that there were a lot of dinky little companies. And their tiny little budgets, when added together, allowed Salesforce.com to improve security, roll out features and services, and attract the likes of Merrill Lynch. Now corporations are switching over at a rapid pace, and the old CRM vendors are sitting around watching the business vanish out from under them.
Today, there are a lot of budget-challenged organizations out there — schools, non-profits, hobbyists — who are using OpenSim to set up free or low-cost grids. Nobody knows how many — OpenSim doesn’t track download numbers (at Hypergrid Business, we only track the big public grids and a couple of specialized distributions of OpenSim.)
And these guys are contributing back to the OpenSim community. They’re sharing content. Organizing events. Reporting bugs — and even donating bug fixes. And they’re paying tiny monthly fees to hosting providers who, in turn, use that money to help improve service and features.
OpenSim is quickly becoming stable enough, robust enough, and full of enough content, events, and features for Second Life’s core customers to start considering switching over. But those customers will continue to be happy with Second Life and prefer it to OpenSim — until the day that they finally decide to leave.
4. Trying to be all things to all people
Like AOL before it, Second Life has everything. It has all the users. It has messaging. It has its own mail system. It has events. It has the server and the viewer, and the payments system and the search and everything else you can think of.
But, just like AOL, Second Life can’t possibly hope to be the best at every single thing.
The World Wide Web disaggregated functions. We now use Google for searching. Facebook and LinkedIn for social networking. Whatever email we prefer for our email. Twitter for sending out 140-character messages. We get our Lolcats from icanhascheezburger.com and our news from CNN.com and our music videos from YouTube. We get our TV from Hulu and our movies from Netflix. We get our books at Amazon and our shoes at Zappos.
We don’t get everything in one place anymore. Instead, we get individual things from the best provider of that particular thing.
Can Linden Lab really be the best possible viewer company and the best possible group messaging company and the best possible events platform and the best possible place for residential homes and the best place for virtual schools and the best place for virtual pets?
Any business strategy based on doing every single thing and doing every single thing well is bound to run into problems at some point.
5. Unhospitable business environment
If you want to trade Second Life’s Linden dollar currency, there are a number of exchanges that will handle the trading, using an application programming interface released by Linden Lab.
European customers, for example, can go to Virwox, and avoid currency conversion fees.
Customers can pick and choose their favorite exchange. Linden Lab, in return, gets to see its currency traded without having to create its own exchanges for the various niche markets.
However, that same kind of openness doesn’t extend to other areas.
For example, corporate customers can’t control their own user accounts — all account administration goes through Linden Lab. By comparison, in OpenSim, each grid handles its own account creation.
Say, for example, I’m a maker of virtual pets and I set up my own OpenSim grid, where all my employees get official company avatars. Customers from all around the hypergrid can teleport in, buy pets, and take them back to their home grid– but only if I decide to let that happen. And, if I want, I can have customers return to my grid when they need to stock up on food, or sell their offspring, or whatever it is that people do with virtual pets.
My grid is fully mine. I own its domain name. I can host it with any hosting company or even on my own servers — just as if it was a website. Nobody can take it away from me or close me down. (Well, maybe the FBI, if I start selling virtual pets to terrorists as part of a money laundering scheme.)
All the server-side software I’ve got running on the grid is mine, and nobody can see it. All the intellectual property is mine — I’m not licensing my own virtual pets back from Second Life. The virtual land is mine. The virtual stores are mine. I can keep my own database of virtual pets and their registered owners, so illegal copies can’t be sold or traded on my grid.
Instead, today, companies doing business inside Second Life are completely dependent on Second Life for everything and they have zero guarantees that they will have the same service tomorrow as they do today. That only works as long as Second Life stays the only game in town.
Something similar is happening with the viewer. Linden Lab did release an open source version of its viewer, and a number of outside developers have used the opportunity to tweak it and come up with their own distributions — Phoenix, Kirsten’s, Hippo, Imprudence, Firestorm, Emergence, Singularity… but none of these viewers are substantially different from the official viewer. In fact, most are created by small teams or individuals adding features that they themselves would like to have in a viewer. Since the third-party viewer developers are all programmers, they tend to add high-end features that other computer-savvy folks enjoy.
But they’re not creating viewers that could be actually useful to growing Second Life — viewers that are easy for newcomers to use, that dramatically and fundamentally change the way the viewer interacts with the virtual environment. Why? Because there’s no profit to it. The Second Life’s viewer code is distributed under a GPL license, which means that companies can’t create proprietary products and charge for them.
By comparison, the OpenSim server code is distributed under a business-friendly BSD license, which means that anyone and everyone create create a sell a custom version of OpenSim. And they do. IBM charges $50,000 for their version of OpenSim integrated with their Lotus Sametime platform.
3Di has a commercial version of OpenSim that’s integrated with their Web-based viewer.
How can Second Life help businesses?
They can release more APIs for third-party viewer developers so that they can build commercial software that ties into Second Life. They could open up search functionality to outside vendors. Messaging. Groups. Account and avatar creation.
Second Life could allow other companies to create businesses on top of their platform, the way that Twitter and Facebook do.
This would create new features and benefits for users, while at the same time allowing Linden Lab more time to focus on improving core functionality.
Second Life is not doomed
Linden Lab doesn’t have to follow in the footsteps of the dying disk drive dinosaurs of The Innovator’s Dilemma.
After all, IBM successfully transitioned from a hardware company to being a leader in global technology services, and there are plenty of other companies that embraced change and reinvented themselves.
Linden Lab can do the same — and I’ve got a few ideas how.